ARTICLE

pubs.acs.org/ac

Semi-Infinite Linear Diffusion Spectroelectrochemistry on

an Aqueous Micro-Drop
Cynthia A. Schroll,† Sayandev Chatterjee,‡ William R. Heineman,*,† and Samuel A. Bryan*,‡
†
Department of Chemistry, University of Cincinnati, Cincinnati, Ohio 45221-0172, United States
‡
Energy and Environment Directorate, Pacific Northwest National Laboratory, Richland, Washington 99352, United States

ABSTRACT: We report a technique for conducting semi-

infinite diffusion spectroelectrochemistry on an aqueous micro-
Downloaded via CENTRO TECNOL ELECTROQUMICA SC on September 26, 2019 at 16:07:19 (UTC).

drop as an easy and economic way of investigating spectro-

See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

electrochemical behavior of redox active compounds and

correlating spectroscopic properties with thermodynamic
potentials on a small scale. The chemical systems used to
demonstrate the aqueous micro-drop technique were an absor-
bance based ionic probe [Fe(CN)6]3
/4
and an emission
based ionic probe [Re(dmpe)3]2þ/þ. These chemical systems
in a micro-drop were evaluated using cyclic voltammetry and UV
visible absorbance and luminescence spectroscopies.

immunoassay on a 50 μL drop sandwiched between the RDE and

O ver the past several decades spectroelectrochemistry has
gained importance as a technique for measuring the redox1
and spectral properties of inorganic,2
7 organic,6,8,9 and biolo-
gical10
13 species. Spectroelectrochemistry can be implemented
in cells that are characterized by either semi-infinite linear
diffusion (SILD) in a bulk solution or by thin layer cells in which
complete electrolysis is rapidly achieved in a thin layer of solution
with restricted diffusion. The spectroscopic methods that have
been used rely on light propagation through the sample based on
transmission, specular reflection, or internal reflection.14 In
transmission spectroscopy the light beam is passed through an
optically transparent electrode and the sample.1,15
17 In specular
reflection spectroscopy the light is passed through the sample,
reflected off the electrode surface, and then passed back through
the sample.18
21 In internal reflection spectroscopy the light is
directed into a waveguide at an angle greater than the critical
angle so that it is internally reflected. At each reflection point an
evanescent electromagnetic wave interacts with the sample, and
the changes in the spectra can be measured.22,23
SILD spectroelectrochemistry has been used for a wide range
of applications.24
27 In spite of the broad scope of SILD spec-
troelectrochemistry, wide applicability of the technique is limited
by cell designs that are complicated to construct and use and
which require relatively large sample sizes—on the order of milli-
liters.15,25,28,29 Cell designs that have the working electrode
positioned so that either thin-layer or SILD spectroelectrochem-
istry can be achieved on a smaller volume of sample have been
reported.30
32
In an effort to substantially reduce the sample size and the cell
complexity of the SILD cell without compromising a rapid
experimental execution time, we herein describe a technique of
micro-drop spectroelectrochemical detection based on a modi-
fication of a method reported by Wijayawardhana et al.33 This
group used a rotating disk electrode (RDE) for electrochemical
an inert bottom plate, with the reference and auxiliary electrodes
inserted into the drop from the side. Our aim is to couple this
electrochemical technique with simultaneous spectroscopic detec-
tion to acquire spectroelectrochemical data in an easy and rapid
way and to keep the total sample volume as small as possible.
This technique would be useful for studying spectroelectro-
chemical processes where minimizing sample size is critically
important because of expense, limited supply, or waste disposal
issues as with radioactive compounds. This work was stimulated
by our particular interest to do spectroelectrochemistry in ionic
liquid solutions of actinides and other radioactive materials and
waste forms where all of these issues apply.
’ EXPERIMENTAL SECTION
Chemicals and Materials. Potassium ferricyanide (K3[Fe(CN)6]),
sodium ferrocyanide (Na4[Fe(CN)6]), and potassium nitrate
(KNO3) were obtained from Sigma-Aldrich and used without
further purification. Tris-(bis-(1,2-dimethylphosphinoethyl))-
rhenium(I) triflate ([Re(dmpe)3](OTf)) was synthesized via
a literature method.34 Aqueous solutions of K3[Fe(CN)6],
Na4[Fe(CN)6], and [Re(dmpe)3](OTf) were made by dissol-
ving the appropriate amount of reagent in 0.1 M KNO3 solution.
To make the glass slide’s surface hydrophobic it was siliconized
with Surfasil (Pierce Chemical Co.). Siliconization was done to
ensure the formation of a spherical drop to allow a sufficiently
large distance between the electrode surface and the glass slide
to have semi-infinite diffusion on the time scale of a typical scan
time for cyclic voltammetry.
Received: March 2, 2011
Accepted: April 16, 2011
Published: April 17, 2011

r 2011 American Chemical Society 4214 dx.doi.org/10.1021/ac200551n | Anal. Chem. 2011, 83, 4214–4219
Analytical Chemistry
Figure 1. (A) Schematic representation of the spectroelectrochemical setup using an aqueous micro-drop, (B) a photograph of the cell. (W = working
electrode, ref = reference electrode, Aux = auxiliary electrode, TP = top plate, QS = quartz slide, BP = bottom plate, RP = reflectance probe).
Instrumentation. Cyclic voltammetry on the aqueous micro-
drop was carried out using an Epsilon Potentiostat (Bioanalytical
Systems) with a standard three-electrode cell stand from Bio-
analytical Systems. All scans were recorded using a glassy carbon
working electrode (area 7.07 mm2) and a Pt wire auxiliary
electrode. Between measurements, the working electrode was
polished with 0.05 μm alumina slurry and a polishing pad, rinsed
with distilled water, and wiped dry using a Kimwipe. All reported
potentials are referenced versus a platinum wire quasi-reference
electrode. Peak currents (ip) were measured with respect to the
extrapolated baseline current as described by Kissinger and
Heineman.35 UV
visible absorption spectra were recorded with
a deuterium light source (Mikropack, model# DH 2000) and an
Ocean Optics USB2000 detector (188
880 nm) using Spectra
Suite Software for spectral data acquisitions. Emission spectra
were recorded using a 532 nm laser excitation source (Melles
Griot, 20 mW CW), coupled to an InSpectrum 150 spectro-
meter-CCD, and using SpectraSense data acquisition software. A
532 nm holographic notch filter (Kaiser) was used to reduce laser
light backscattered into the InSpectrum 150 spectrometer. Signal
integration times were typically 500 ms using a 2-mm slit width
for a 600 gr/mm grating blazed at 500 nm. Step-index silica-on-
silica optical fibers were purchased from Romack, Inc. The Ocean
Optics system consisted of a USB-200FL spectrometer and
Ocean Optics 00IBase32 Spectroscopy Software.
In a typical setup as illustrated in Figure 1, a glass slide was
placed on a stand above the reflectance probe. The custom
designed stand consisted of two aluminum plates separated by an
O-ring spacer to allow an optically transparent surface (e.g.,
quartz plate or glass microscope slide) to slide in. A hole was
drilled at the center of the aluminum plates to permit passage of
the optical beam. The upper aluminum plate was held in place
with thumb screws that also provided conducting surfaces to
attach the reference and working electrodes. The auxiliary and
reference wire electrodes were spaced approximately 1 mm apart
from one another. To ensure that the electrodes were relatively
immobile during the entire course of the experiment, they were
glued to the posts with epoxy resin. This limited the free mobility
of the reference and the auxiliary electrodes, and preserved the
relative distance, alignment, and orientation between the indivi-
dual electrodes. The working electrode was lowered to within
∼1 mm from the glass surface, and a 30 μL drop of solution was
placed in between. The distance of the tip of the working elec-
trode from the glass surface was set at 1 mm (which equals half
the total optical path length) using slide calipers. The path length
4215
ARTICLE
determined from the absorbance at 420 nm of a microdrop of a
solution containing a known concentration of K3Fe(CN)6, con-
firmed the distance to within 5% of experimental error. The drop
was positioned to cover the reference and auxiliary electrodes
while surface tension ensured its contact with the working electrode
from above. A 180° back-reflectance UV
visible-NIR probe
with a bundle of fiber-optic cables was fixed to a clamp directly
below the glass slide surface for spectral measurement and
focused directly on the surface of the working electrode. The
reflectance probe configuration was six peripheral fiber optic
cables for excitation coming from the light source, with one
concentrically located cable for collection of the light trans-
mitted after reflection from the surface of the disk working
electrode, going to the detector.
’ RESULTS AND DISCUSSION
To explore this technique, two model compounds with known
electrochemical and spectroscopic properties were chosen. The
chemical systems used to demonstrate the micro-drop technique
were an absorbance based ionic probe [Fe(CN)6]3
/4
and a
luminescence based ionic probe [Re(dmpe)3]2þ/þ. Both sys-
tems are characterized by a chemically reversible electrochemical
couple and demonstrate the general requirements of absor-
bance and luminescence based systems, respectively.
[Fe(CN)6]3
/[Fe(CN)6]4
System. For absorbance experiments,
the [Fe(CN)6]3
/[Fe(CN)6]4
couple (E°0 = 0.25 V vs Ag/AgCl)
was chosen. The two redox forms are spectroscopically distinguish-
able; an aqueous solution of [Fe(CN)6]4
(pH = 7, T = 25 °C) is
optically transparent at 420 nm while [Fe(CN)6]3
has a strong
absorption band at this wavelength under identical conditions
(ε420 = 1040 M
1 cm
1).36 Thus, oxidation of Fe(II) to Fe(III)
results in the appearance of the 420 nm band, while the reduction
of [Fe(CN)6]3
back to [Fe(CN)6]4
results in a decrease in its
intensity.
Cyclic Voltammetry of [Fe(CN)6]3
/[Fe(CN)6]4
. Cyclic vol-
tammetry was used to evaluate the electrochemical properties of
the cell, measure E°0 , and determine the diffusion coefficients of
the [Fe(CN)6]3
/[Fe(CN)6]4
system. The E°0 for this redox
process was
0.1 V vs the quasi-reference electrode for all scans
with a ΔEp = 120 mV (ν = 50 mV s
1). This ΔEp is larger than
the expected 59 mV for an electrochemically reversible 1e

system and is attributed to the slow electron transfer that can
occur at glassy carbon. Cyclic voltammograms of 5.0 mM K3-
[Fe(CN)6] in 0.1 M KNO3 solution as a function of scan rate
dx.doi.org/10.1021/ac200551n |Anal. Chem. 2011, 83, 4214–4219
Analytical Chemistry
Figure 2. (A) Cyclic voltammograms of an aqueous solution of 5.0 
10
3 M K3[Fe(CN)6], 0.1 M KNO3 as a function of scan rate, (light
brown line) 5, (red line) 10, (yellow line)) 25, (light green line) 50,
(green line) 75, and (light blue line) 100 mV/s (GC disk working
electrode, Pt wire reference electrode and Pt wire auxiliary electrode).
(B) Plot of peak current versus the square root of scan rate: (solid light
brown squares) anodic wave; ipa(μA) =
114 ν1/2 (V s
1)1/2
6.1 (R2 =
0.997); (solid blue diamonds) cathodic wave; ipc(μA) = 149 ν1/2 (V s
1)1/2 þ
3.9 (R2 = 0.997).
(Figure 2A) have voltammograms that are consistent with semi-
infinite diffusion for scan rates ranging from 5 to100 mV s
1.37
The positive shift in the potential of both the reduction and the
oxidation peaks is due to the change in concentration of the
oxidative and reductive species with respect to the Pt quasi-
reference electrode. This was further confirmed by a control
experiment where the cyclic voltammetry of a 5.0 mM K3[Fe-
(CN)6] in 0.1 M KNO3 solution in a standard cell referenced to a
Pt quasi reference electrode was compared to that using a Ag/
AgCl reference electrode. The potential for both peaks again
shifted in a positive direction when the Ag/AgCl reference
electrode was replaced by the Pt quasi reference.
A plot of peak currents versus the square root of scan rates
(Figure 2B) is linear as expected for semi-infinite diffusion. Using
the Randle
Sevcik equation35 the diffusion coefficient for
[Fe(CN)6]3
was found to be 2.4  10
6 cm2 s
1 which is in
agreement with the literature.38 The peak current ratio of ip/ic =
0.76 indicates that the [Fe(CN)6]3
/[Fe(CN)6]4
redox
couple is quasi-reversible. The quasi-reversibility is indicative
of adsorption of analyte to the surface of the glassy carbon
from the solution.39 The performance is very similar to that
obtained using a glassy carbon working electrode in a conven-
tional regular volume three component electrochemical cell setup
and confirms the semi-infinite linear diffusion, even at slower
scan rates.
Chronoabsorptometry of [Fe(CN)6]3
/[Fe(CN)6]4
. Double
potential step chronoabsorptometry was used to evaluate the
spectroelectrochemical characteristics of the cell. In a typical
experiment, a forward step of 0.4 V was applied to an aqueous
solution of 25 mM [Fe(CN)6]4
, 0.1 M KNO3 for 90 s while
absorption spectra were concurrently recorded at 5 s intervals.
The oxidation of [Fe(CN)6]4
to [Fe(CN)6]3
was character-
ized by the appearance and growth of the 420 nm absorption
band. Subsequently, the reverse step to
0.4 V for 90 s resulted
in a decrease in intensity of this absorption band, demonstrating a
reduction of [Fe(CN)6]3
back to [Fe(CN)6]4
. A plot of
absorbance versus time at 420 nm for both forward and reverse
potential steps is shown in Figure 3A. It can be seen that the
system levels off to a maximum absorbance within 40 s from the
start of the oxidation while a minimum value is reached within
40 s of the start of the back reduction. Using the equation:
δ ¼ ð2DtÞ1=2
ARTICLE
Figure 3. (A) Double potential step chronoabsorptometry at 420 nm of
an aqueous solution of 25.0  10
3 M Na4[Fe(CN)6], 0.1 M KNO3
(Glassy carbon working electrode, Pt wire reference electrode and Pt
wire auxiliary electrode). A potential of 0.4 V applied for 90 s resulted in
an increase and subsequent leveling off in absorbance. An immediate
subsequent application of
0.4 V resulted in a decrease followed by
subsequent leveling off of the absorbance. (B) Plot of Absorbance vs
Time1/2, for the oxidation step. The red squares represent the diffusion
controlled region, while the blue squares represent the deviation. The
equation for the diffusion controlled region is A(a.u.) = 0.047 t1/2(s1/2) þ
0.02 (R2 = 0.98). The standard errors for the slope and intercept are
(þ0.002,
0.007) and ((0.006), respectively.
where δ is the Nernst diffusion layer, D is the diffusion coefficient
of ferrocyanide, and t is time, it is calculated that the electrolysis
should take about 60 min. This calculated time for complete
electrolysis confirms that under our experimental parameters,
where each potential was held for less than 2 min, only a small
fraction of the total volume of the working solution was electro-
lyzed, and semi-infinite diffusion conditions existed.
Thus, the leveling off of absorbance in the forward step of
Figure 3A is not due to complete electrolysis of the absorbing
species, but rather is attributed to the optical configuration of the
cell. The reflectance probe surface, composed of a fiber bundle
2 mm in diameter, illuminates the glassy carbon electrode (GCE),
whose surface area is 3 mm in diameter, located a distance of
1 mm away from each other. At this distance, the reflectance
probe beam divergence is larger than the active area of the GCE,
and under the existing experimental settings, the spectroscopi-
cally interrogated solution volume includes solution outside of
the diffusion layer adjacent to the GCE. Under these conditions,
Beer’s Law behavior will not strictly be observed, which explains
why the change in absorbance in Figure 3A levels off more rapidly
than otherwise would be predicted. Using a larger electrode
surface area or a more collimated beam would reduce this effect.
Additionally, by continuing the measurement for longer than 5 min
the gradual evaporation of the drop was observed, as evidenced by
an increase in the absorbance intensity of the plateau even while
ð1Þ holding the potential of the working electrode at
0.4 V.
4216 dx.doi.org/10.1021/ac200551n |Anal. Chem. 2011, 83, 4214–4219
Analytical Chemistry
Figure 4. (A) Cyclic voltammogram of an aqueous solution of 1.25 
10
3 M [Re(dmpe)3](OTf), 0.1 M KNO3 at ν = 50 mV s
1 showing the
two redox couples of Re(dmpe)3(OTf). (B) Cyclic voltammograms of
an aqueous solution of 1.25  10
3 M Re(dmpe)3(OTf), 0.1 M KNO3
as a function of scan rate, (light brown line) 5, (red line) 10, (yellow line)
15, (light green line) 25, (green line) 36, (light blue line) 50, (blue line)
65, and (dark blue line) 80 mV/s (Glassy carbon working electrode, Pt
wire reference electrode and Pt wire auxiliary electrode). (C) Plot of
peak current versus the square root of scan rate: (solid red squares)
anodic process for [Re(dmpe)3]2þ/[Re(dmpe)3]þ; ipa (μA) =
15.6
ν1/2 (V s
1)1/2
0.8 (R2 = 0.997); (solid blue diamonds) cathodic
process for [Re(dmpe)3]2þ/[Re(dmpe)3]þ; ipc (μA) = 14.1 ν1/2
(V s
1)1/2 þ 0.005 (R2 = 0.994).
A plot of absorbance versus the square root of time for the
oxidation process is shown in Figure 3B. From this plot it can be
seen that the absorbance change starts as a diffusion controlled
system, but at approximately 40 s there is a deviation from the
diffusion controlled system for the reasons discussed above. Applying
the equation for chronoabsorptometry to the slope of the linear
portion of the graph (Figure 3B) that adheres to Beer’s Law:
2ε0 D1=2 Ct 1=2
A¼
π1=2
where A is the absorbance, εo is the molar absorptivity of ferricyanide,
D is defined from eq 1, C is the concentration of ferrocyanide,
and t is time; the molar absorptivity of ferricyanide was
calculated be ε420 nm = 1390 M
1 cm
1.40,41 This is similar
to the accepted value of 1100 M
1 cm
1 42 and confirms the
semi-infinite diffusion conditions.
[Re(dmpe)3]2þ/[Re(dmpe)3]þ System. For emission experi-
ments, [Re(dmpe)3](OTf) was chosen to take advantage of the
changes in luminescence properties of the complex associated
with the reversible [Re(dmpe)3]2þ/þ (E°0 = 0.34 V vs Ag/AgCl)
redox couple. [Re(dmpe)3]2þ emits at 605 nm (λex = 532 nm)
while [Re(dmpe)3]þ does not emit at this wavelength.43 The
complex has also been reported to exhibit a second redox
process (E°0 = 1.12 V vs Ag/AgCl; chemically quasi-reversible
as determined by the ratio of the peak currents) assigned to the
Re3þ/2þ couple.
Cyclic Voltammetry of [Re(dmpe)3 ]2þ/[Re(dmpe)3 ]þ.
Cyclic voltammetry was used to determine the E°0 of both
the [Re(dmpe)3 ]2þ/[Re(dmpe)3]þ and [Re(dmpe)3]3þ/
[Re(dmpe)3]2þ couples as well as the diffusion coefficients for
[Re(dmpe)3]2þ and [Re(dmpe)3]þ. The cyclic voltammograms
of the complex within the potential range 1.2 V and
0.3 V
ARTICLE
shows the two redox processes (Figure 4A). The lower potential
process (E°0 = 0.002 V; ΔEp = 87 mV) is assigned to the
[Re(dmpe)3]2þ/[Re(dmpe)3]þ couple while the higher potential
process (E°0 = 0.97 V; ΔEp = 71 mV) is assigned to the Re3þ/2þ
couple . Cyclic voltammograms as a function of scan rate for the
[Re(dmpe)3]2þ/[Re(dmpe)3]þ couple (Figure 4B) have voltam-
mograms that are consistent with semi-infinite diffusion for scan
rates from 5 to 80 mV s
1.43 As expected, the increase in peak
current is directly related to the increase in the square root of scan
rate (Figure 4C). The ratio of the slope of the peak currents as a
function of square root of scan rate for the [Re(dmpe)3]2þ/
[Re(dmpe)3]þ couple is calculated to be ip/ic = 0.90. It was
expected that the ip/ic would be less than one because of the air
oxidation of [Re(dmpe)3]þ to [Re(dmpe)3]2þ. Furthermore, a
Randle
Sevcik analysis35 showed that [Re(dmpe)3]þ has a diffu-
sion coefficient of 4.34  10
7 cm2 s
1. The diffusion coefficient of
[Re(dmpe)3]2þ during its reduction to [Re(dmpe)3]þ is 3.53 
10
7 cm2 s
1. This performance is close to that obtained for a glassy
carbon electrode dipped into a conventional electrochemical cell
confirming semi-infinite diffusion, even at the slower scan rates.
Because of significantly reduced chemical reversibility and signifi-
cantly less pronounced changes in the spectroscopic properties, the
[Re(dmpe)3]3þ/[Re(dmpe)3]2þcouple was not further explored
for the purpose of this paper.
Luminescence Based Spectroelectrochemistry of
[Re(dmpe)3]2þ/[Re(dmpe)3]þ. The cell’s capacity for lumines-
cence based spectroelectrochemistry was demonstrated using the
[Re(dmpe)3]2þ/[Re(dmpe)3]þ couple. Since [Re(dmpe)3]þ
air oxidizes to [Re(dmpe)3]2þ, a water-saturated N2 blanket
was placed over the working electrode assembly. A potential
of 0.4 V was applied to an aqueous solution of 1.25 mM
[Re(dmpe)3](OTf), 0.1 M KNO3 for 90 s, and a series of
luminescence spectra were recorded at 10 s intervals under
532 nm laser excitation. The oxidation of [Re(dmpe)3]þ to
[Re(dmpe)3]2þ was characterized by the appearance and growth
of the 605 nm emission band. An immediate change in potential
to
0.6 V for 90 s resulted in a decrease in intensity of the 605 nm
emission band, indicating reduction of [Re(dmpe)3]2þ back to
ð2Þ
the starting material, [Re(dmpe)3]þ (Figure 5A).
The profile of the emission maximum is shown in the time
versus 605 nm emission intensity plot (Figure 5B). Modula-
tion of luminescence intensity at 605 nm by sequential
application of a positive potential (0.4 V) for 120 s, followed
by immediate application of a negative potential (
0.6 V)
repeated for two cycles, showed a sinusoidal behavior in the
intensity. The baseline slope (the line formed by extrapolation
through the local intensity minima) showed a positive slope of
1.3. A control experiment conducted in which luminescence
intensity of an aqueous solution of 1.25 mM [Re(dmpe)3]-
(OTf), 0.1 M KNO3 was monitored under excitation by the
same 532 nm laser source in the absence of any potential
exhibited a similar slope (1.5), indicating the slope in the
modulated curve is due to aerial oxidation. Figure 5B shows the
intensity vs time for the control run, illustrating that in the
absence of an applied potential, there is a gradual appearance of
the 605 nm emission band. We are currently working on
modifications to reduce exposure to atmosphere to limit the
chemical oxidation of air sensitive analytes. Shown in
Figure 5C is a plot of luminescence vs time1/2 for the first
oxidation of [Re(dmpe)3]þ. The linear plot confirms semi-
infinite diffusion throughout the step.
4217 dx.doi.org/10.1021/ac200551n |Anal. Chem. 2011, 83, 4214–4219
Analytical Chemistry ARTICLE

Figure 5. (A) Modulation of luminescence intensity of an aqueous solution 1.25  10
3 M [Re(dmpe)3](OTf), 0.1 M KNO3 with time on the
application of voltage (λ ex = 532 nm). A potential of 0.4 V applied for 120 s resulted in an increase and subsequent leveling off in intensity. An immediate
subsequent application of
0.6 V for 120 s resulted in a decrease followed by subsequent leveling off of the intensity. This process was repeated for two
cycles. (B) Modulation of luminescence intensity at 605 nm: (solid blue squares) Points at the trough of each modulation where there is a maximum
concentration of [Re(dmpe)]þ present. The equation of the line passing through the troughs is I(a.u.) = 1.60t(s) þ 360. (solid light bown squares)
Control run of solution 1.25  10
3 M [Re(dmpe)3](OTf), 0.1 M KNO3 where no voltage was applied. The equation of the line is I(a.u.) = 1.50t(s) .
(C) Plot of Luminescence vs Time1/2 for the oxidation of 1.25  10
3 M [Re(dmpe)3](OTf), 0.1 M KNO3. The equation of the line is I(a.u.) = 101t1/
(s ) þ 9 (R2 = 0.98). The standard errors for the slope and intercept are (þ 1,
6) and (( 10), respectively.
2 1/2

’ CONCLUSIONS ’ AUTHOR INFORMATION

SILD spectroelectrochemistry using either absorption or
emission spectroscopy can be done on a 30 μL drop of liquid
with a cell that is easy to fabricate and use. SILD conditions were
maintained over the time scale corresponding to cyclic voltam-
metry scan rates as slow as 2 mV s
1. Optical sensitivity was
sufficient to easily monitor absorbance and fluorescence changes
associated with the electrochemistry. The cell has the advantage
of not requiring electrode transparency, making it more widely
applicable than spectroelectrochemical techniques in which
optical measurements are made through the electrode. The cell
could be easily converted into a thin layer cell by simply reducing
the distance between the electrode and the bottom plate. Improve-
ments that could be made include enlarging the electrode surface
or reducing the optical beam so that only the diffusion layer is
monitored and provision for deoxygenation. The effectiveness and
reproducibility of this method for studying spectroelectrochemical
processes in microquantities of analyte makes the technique
promising for substances that are expensive, limited in available
quantity, or pose disposal issues because of toxicity or radioactivity.
4218
Corresponding Author
*E-mail: sam.bryan@pnl.gov (S.A.B.), heinemwr@ucmail.uc.edu
(W.R.H.).
’ ACKNOWLEDGMENT
This research was supported by the U.S. Department of
Energy's Fuel Cycle Research and Development (FCR&D),
Separation Campaign (NE) and performed at the Pacific North-
west National Laboratory operated by Battelle for the U.S.
Department of Energy under Contract DE-AC05-76RL01830.
Support from the Office of Science, BER (Grant No. DE-FG02-
07ER64353), and the Office of Environmental Management
Sciences Program (Grant No. DE-FG0799ER62331) of the U.S.
Department of Energy (DOE) is also gratefully acknowledged.
We thank Dr. Necati Kaval for designing the InSpectrum 150
spectrometer-CCD, Eme Amba Abu for helpful discussions, and
Amos Doepke for preparing the hydrophobic glass slides.
dx.doi.org/10.1021/ac200551n |Anal. Chem. 2011, 83, 4214–4219
Analytical Chemistry ARTICLE

’ REFERENCES
(1) Kuwana, T.; Darlington, R. K.; Leedy, D. W. Anal. Chem. 1964,
36, 2023.
(2) Rieder, K.; Hauser, U.; Siegenthaler, H.; Schmidt, E.; Ludi, A.
Inorg. Chem. 1975, 14, 1902.
(3) Rohrbach, D. F.; Heineman, W. R.; Deutsch, E. Inorg. Chem.
1979, 18, 2536.
(4) Hill, M. G.; Penneau, J. F.; Zinger, B.; Mann, K. R.; Miller, L. L.
Chem. Mater. 1992, 4, 1106.
(5) Kaim, W.; Klein, A.; Hasenzahl, S.; Stoll, H.; Zalis, S.; Fiedler, J.
Organometallics 1998, 17, 237.
(6) Toma, H. E.; Araki, K. Curr. Org. Chem. 2002, 6, 21.
(7) Lahiri, G. K.; Kaim, W. Dalton Trans. 2010, 39, 4471.
(8) Owens, J. L.; Marsh, H. A.; Dryhurst, G. J. Electroanal. Chem.
1978, 91, 231.
(9) Stutts, K. J.; Powell, L. A.; Wightman, R. M. J. Electrochem. Soc.
1981, 128, 1248.
(10) Heineman, W. R.; Norris, B. J.; Goelz, J. F. Anal. Chem. 1975,
47, 79.
(11) Sailasuta, N.; Anson, F. C.; Gray, H. B. J. Am. Chem. Soc. 1979,
101, 455.
(12) Larsson, T.; Lindgren, A.; Ruzgas, T. Bioelectrochemistry 2001,
53, 243.
(13) Comtat, M.; Durliat, H. Biosens. Bioelectron. 1994, 9, 663.
(14) Heineman, W. R. Anal. Chem. 1978, 50, 390A.
(15) Strojek, J. W.; Kuwana, T. J. Electroanal. Chem. 1968, 16, 471.
(16) Kuwana, T.; Strojek, J. W. Faraday Discuss. 1968, 45, 134.
(17) Osa, T.; Kuwana, T. J. Electroanal. Chem. 1969, 22, 389.
(18) Bewick, A.; Tuxford, A. M. Symp. Faraday. Soc. 1970, 4, 114.
(19) Aylmer-Kelly, A. W. B.; Bewick, A.; Cantrill, P. R.; Tuxford,
A. M. Faraday Discuss. 1973, 56, 96.
(20) Pruiksma, R. J. Electroanal. Chem. 1980, 114, 147.
(21) Gaillard, F.; Levillain, E. J. Electroanal. Chem. 1995, 398, 77.
(22) Hansen, W. N.; Kuwana, T.; Osteryoung, R. A. Anal. Chem.
1966, 38, 1810.
(23) Prostak, A.; Mark, H. B.; Hansen, W. N. J. Phys. Chem. 1968,
72, 2576.
(24) Kuwana, T.; Heineman, W. R. Acc. Chem. Res. 1976, 9, 241.
(25) Kuwana, T.; Winograd, N. In Electroanalytical Chemistry; Bard,
A. J., Ed.; Marcel Dekker: New York, 1974; p 1.
(26) Heineman, W. R.; Hawkridge, F. M.; Blount, H. N. In Electro-
analytical Chemistry; Bard, A. J., Ed.; Marcel Dekker: New York, 1984;
p 1.
(27) Niu, J.; Dong, S. Rev. Anal. Chem. 1996, 15, 1.
(28) Winograd, N.; Kuwana, T. J. Electroanal. Chem. 1969, 23, 333.
(29) Winograd, N.; Kuwana, T. Anal. Chem. 1971, 43, 252.
(30) Porter, M. D.; Dong, S. J.; Gui, Y. P.; Kuwana, T. Anal. Chem.
1984, 56, 2263.
(31) Salbeck, J. J. Electroanal. Chem. 1992, 340, 169.
(32) Heineman, W. R.; Kuwana, T. Anal. Chem. 1972, 44, 1972.
(33) Wijayawardhana, C. A.; Halsall, H. B.; Heineman, W. R. Anal.
Chim. Acta 1999, 399, 3.
(34) Deutsch, E.; Libson, K.; Vanderheyden, J. L.; Ketring, A. R.;
Maxon, H. R. Int. J. Radiat. Appl. Instrum. B 1986, 13, 465.
(35) Kissinger, P. T.; Heineman, W. R. J. Chem. Educ. 1983, 60, 702.
(36) Krenitsky, T. A.; Neil, S. M.; Elion, G. B.; Hitchings, G. H. Arch.
Biochem. Biophys. 1972, 150, 585.
(37) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Funda-
mentals and Applications, 2nd ed; Wiley: New York, 2001; p 147.
(38) Konopka, S. J.; McDuffie, B. Anal. Chem. 1970, 42, 1741.
(39) Yang, R.; Qian, Z.; Deng, J. J. Electrochem. Soc. 1998, 145, 2231.
(40) Ogura, K.; Naito, T. Electrochim. Acta 1982, 27, 1243.
(41) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Funda-
mentals and Application, 2 ed.; Wiley: New York, 2001; p 682.
(42) Alexander, J. J.; Gray, H. B. J. Am. Chem. Soc. 1968, 90, 4260.
(43) Kirchhoff, J. R.; Allen, M. R.; Cheesman, B. V.; Okamoto, K.;
Heineman, W. R.; Deutsch, E. Inorg. Chim. Acta 1997, 262, 195.

4219 dx.doi.org/10.1021/ac200551n |Anal. Chem. 2011, 83, 4214–4219